Accepted Manuscript

Title: Small Bowel Protection against NSAID-injury in Rats:Effect of Rifaximin, a poorly absorbed, GI targeted, antibiotic

Author: Matteo Fornai Luca Antonioli Caronina PellegriniRocchina Colucci Deborah Sacco Erika Tirotta GianfrancoNatale Alessia Bartalucci Marina Flaibani Cecilia RenzulliEmilia Ghelardi Corrado Blandizzi Carmelo Scarpignato

PII: S1043-6618(15)00296-0DOI: http://dx.doi.org/doi:10.1016/j.phrs.2015.12.031Reference: YPHRS 3024

To appear in: Pharmacological Research

Received date: 8-8-2015Revised date: 17-12-2015Accepted date: 25-12-2015

Please cite this article as: Fornai Matteo, Antonioli Luca, Pellegrini Caronina,Colucci Rocchina, Sacco Deborah, Tirotta Erika, Natale Gianfranco, BartalucciAlessia, Flaibani Marina, Renzulli Cecilia, Ghelardi Emilia, Blandizzi Corrado,Scarpignato Carmelo.Small Bowel Protection against NSAID-injury in Rats: Effectof Rifaximin, a poorly absorbed, GI targeted, antibiotic.Pharmacological Researchhttp://dx.doi.org/10.1016/j.phrs.2015.12.031

This is a PDF file of an unedited manuscript that has been accepted for publication.As a service to our customers we are providing this early version of the manuscript.The manuscript will undergo copyediting, typesetting, and review of the resulting proofbefore it is published in its final form. Please note that during the production processerrors may be discovered which could affect the content, and all legal disclaimers thatapply to the journal pertain.

1

Small Bowel Protection against NSAID-injury in Rats: Effect of

Rifaximin, a poorly absorbed, GI targeted, antibiotic

Matteo Fornaia, Luca Antoniolia, Caronina Pellegrinia, Rocchina Coluccib, Deborah

Saccoa, Erika Tirottaa, Gianfranco Natalec, Alessia Bartaluccic, Marina Flaibanic, Cecilia

Renzullid, Emilia Ghelardie, Corrado Blandizzia, Carmelo Scarpignatof

aDivision of Pharmacology, Department of Clinical & Experimental Medicine, University of Pisa, Via Roma 55, 56126 Pisa, Italy

bDepartment of Pharmaceutical and Pharmacological Sciences, University of Padova,

Padova, Italy cDepartment of Translational Research and New Technologies in Medicine and Surgery, University of Pisa, Via Roma 55, 56126 Pisa, Italy

dDepartment of Research & Development, Alfa Wassermann SpA, Via Ragazzi del ‘99,

5, 40133, Bologna, Italy

eDepartment of Translational Research and New Technologies in Medicine and

Surgery, University of Pisa Via San Zeno 37, 56127 Pisa

fClinical Pharmacology & Digestive Pathophysiology Unit, Department of Clinical &

Experimental Medicine, University of Parma, Via Gramsci 14, 43126, Parma, Italy

Address for Correspondence:

Carmelo SCARPIGNATO, MD, DSc (Hons), PharmD, MPH, FRCP (London), FACP, FCP, FACG, AGAF

Professor of Pharmacology & Therapeutics

Associate Professor of Gastroenterology

Head, Clinical Pharmacology & Digestive Pathophysiology Unit

Department of Clinical & Experimental Medicine, University of Parma

Maggiore University Hospital, Cattani Pavillon, I-43125 Parma, Italy

Phone: +39 0521 903863; E-mail: scarpi@tin.it

2

Graphical Abstract

3

Abstract

Nonsteroidal anti-inflammatory drugs, besides exerting detrimental effects on the

upper digestive tract, can also damage the small and large intestine. Although the

underlying mechanisms remain unclear, there is evidence that enteric bacteria play a

pivotal role. The present study examined the enteroprotective effects of a delayed-

release formulation of rifaximin-EIR (R-EIR, 50 mg/kg BID, i.g.), a poorly absorbed

antibiotic with a broad spectrum of antibacterial activity, in a rat model of enteropathy

induced by indomethacin (IND, 1.5 mg/kg BID for 14 days) administration. R-EIR was

administered starting 7 days before or in concomitance with IND administration. At

the end of treatments, blood samples were collected to evaluate hemoglobin (Hb)

concentration (as an index of digestive bleeding). Small intestine was processed for: 1)

histological assessment of intestinal damage (percentage length of lesions over the

total length examined); 2) assay of tissue myeloperoxidase (MPO) and TNF levels, as

markers of inflammation; 3) assay of tissue malondialdehyde (MDA) and protein

carbonyl concentrations, as an index of lipid and protein peroxidation, respectively; 4)

evaluation of the major bacterial phyla. IND significantly decreased Hb levels, this

effect being significantly blunted by R-EIR. IND also induced the occurrence of lesions

in the jejunum and ileum. In both intestinal regions, R-EIR significantly reduced the

percentage of lesions, as compared with rats receiving IND alone. Either the markers of

inflammation and tissue peroxidation were significantly increased in jejunum and

ileum from IND-treated rats. However, in rats treated with R-EIR, these parameters

were not significantly different from those observed in controls. R-EIR was also able to

counterbalance the increase in Proteobacteria and Firmicutes abundance induced by

INDO. R-EIR treatment significantly prevents IND-induced intestinal damage, this

enteroprotective effect being associated with a decrease in tissue inflammation,

oxidative stress and digestive bleeding as well as reversal of NSAID-induced alterations

in bacterial population.

Keywords: Nonsteroidal anti-inflammatory drugs, intestinal damage, intestinal

bleeding, rifaximin, enteroprotection, bacterial flora

4

1. Introduction

Nonsteroidal anti-inflammatory drugs (NSAIDs) are very effective medications [1,2],

but their use is associated with a broad spectrum of adverse reactions involving the

liver, kidney, cardiovascular (CV) system, skin and gut [3]. Gastrointestinal (GI) adverse

effects are the most common and cover a wide clinical spectrum, ranging from

dyspepsia, heartburn and abdominal discomfort to more serious events, such as peptic

ulcer with life-threatening complications of bleeding and perforation [4,5].

While proton pump inhibitors (PPIs) reduce the development of peptic ulcer and

related complications in patients taking NSAIDs or/and low-dose aspirin, their

beneficial effect, which is related to their gastric antisecretory activity [6], is not

expected to take place beyond the duodenum [7]. The appreciation that NSAID-

associated GI damage does extend also to the lower digestive tract dates back to the

early 90’s, when a few observational studies and the first large prevention study (i.e.

the MUCOSA trial) were published [8]. In the more recent VIGOR trial, more than 40%

of NSAID-related events occurred in the lower GI tract (i.e. small bowel and colon) [9].

Over the past decade, there has been a progressive change in the overall pattern of GI

events leading to hospitalization, with a clear decreasing trend in upper GI events and

a slight, but significant, increase in lower GI events [10]. The availability of video

capsule endoscopy has allowed a precise quantification of the incidence and

characterization of small bowel damage, which appears to be time-dependent. Indeed,

available studies [11,12] have shown that about 75% of NSAID users display intestinal

mucosal injury, with most denuded areas identified in the proximal part of the small

bowel and all ulcers in its distal part [13]. In healthy volunteers [12,14,15] and patients

[11], omeprazole did not prevent NSAID-associated intestinal damage, evaluated by

video capsule or/and fecal calprotectin measurement. Recent experimental (for review

see [16]) and clinical [17,18] evidence suggests that PPIs may actually aggravate NSAID

injury in the small bowel. The lack of protective effect is clearly due to the fact that

5

NSAID-enteropathy does not depend on gastric acid and is therefore not a pH-

dependent phenomenon [5].

The pathogenesis of small intestinal damage is still not completely understood.

Although drug-induced inhibition of mucosal prostaglandin (PG) synthesis during

NSAID use occurs along the entire digestive tract, there are significant differences

between the distal and proximal GI tract in the concurrence of other pathogenic

factors that may add to mucosal damage. Among them, the absence of acid and the

presence in the intestinal lumen of bacteria and bile, which may trigger specific NSAID-

related pathogenic mechanisms at level of the distal GI tract, are the most prominent

ones [19].

Increasing experimental evidence suggests that inhibition of both COX-1 and COX-2 is

necessary to cause significant GI damage [20,21,22]. However, NSAID-induced injury to

the intestinal epithelium is set in motion by direct effects of the drug after oral

administration, a persistent local action, due to enterohepatic circulation and systemic

effects after absorption. Initial cellular damage is due to entrance of the usually acidic

NSAIDs into the cell via damage to the brush border cell membrane, and disruption of

the mitochondrial processes of oxidative phosphorylation, with consequent ATP

deficiency [20,21,22]. This leads to increased mucosal permeability [23], which

facilitates the entry and actions of luminal factors, such as dietary macromolecules,

bile acids, components of pancreatic juice, and bacteria, activating the inflammatory

cascade [20,21,22].

Amongst luminal aggressors, intestinal bacteria are the main neutrophil

chemoattractants. Several studies [24] show that antimicrobials (tetracycline,

kanamycin, metronidazole or neomycin plus bacitracin) attenuate NSAID-enteropathy,

thus supporting further the pathogenic role of enteric bacteria. Additional, albeit

indirect, support to the role of gut bacteria in the pathogenesis of NSAID-enteropathy,

is represented by the similarity between indomethacin-induced intestinal damage and

6

Crohn’s disease (CD). Not only are the lesions both macro- and microscopically similar,

but are they also sensitive to the same drugs (e.g. sulphasalazine, corticosteroids,

immunosuppressants and antibiotics) [25], at least in the experimental setting. A

recent video capsule study [26] showed also that mesalazine granules were able to

reduce naproxen-induced intestinal damage in patients with inflammatory

arthropaties.

Early studies [27] showed that metronidazole (an antimicrobial targeting most Gram-

negative and Gram-positive anaerobic bacteria [28]) is able to reduce inflammation

and blood loss in patients taking NSAIDs, thus suggesting a therapeutic potential of

antimicrobials in this clinical setting. However, potential adverse effects of systemic

antimicrobials and the possible occurrence of drug resistance have so far precluded

this interesting approach [24].

Rifaximin (4-deoxy-4'-methylpyrido[1',2'-1,2]imidazo [5,4-c]rifamycin SV) is a synthetic

derivative of rifamycin, characterized by very low GI absorption, while retaining a

broad spectrum of antibacterial activity [29,30,31]. Being virtually non-absorbed, its GI

bioavailability is high, with fecal concentrations largely exceeding minimum inhibitory

concentrations against pathogenic enterobacteria, while its scarce impact on extra-GI

sites minimizes the risk of antimicrobial resistance and systemic adverse events [25].

Upon appreciation of the pathogenic role of gut flora in several GI diseases, the use of

rifaximin has been extended from GI infections to hepatic encephalopathy, small

intestine bacterial overgrowth and colonic diverticular disease [25,32]. The drug is also

being investigated in the treatment of inflammatory bowel disease [33].

Since rifaximin does display all the characteristics of an ideal antibiotic for targeting

enterobacteria [34], we felt it worthwhile to assess its ability to prevent NSAID-

enteropathy in a rat model of indomethacin-induced enteropathy. To this aim, we

selected the recently developed extended intestinal release (EIR) formulation, which

contains microgranules of rifaximin, coated with a gastric acid-resistant polymer. This

7

formulation, under development also for the treatment of Crohn’s disease, has been

designed to bypass the stomach and release the microgranules directly in the

intestine, thereby increasing the local concentration of rifaximin, and to maximize its

therapeutic efficacy [33].

2. Methods

2.1 Animals

Experiments were performed on aged (40 week-old) male Wistar rats (500-600 g)

(Harlan Laboratories, Udine, Italy). The animals were fed standard laboratory chow

and tap water ad libitum, and were not employed for at least one week after their

delivery to the laboratory. They were housed, three in a cage, in temperature-

controlled rooms on a 12-h light cycle at 22-24°C and 50-60% humidity.

2.2 Experimental design

Enteropathy was induced by indomethacin in accordance with the method previously

developed in our laboratory [35]. The dose of indomethacin and duration of treatment

were selected in order to obtain small bowel injuries mirroring those induced by

NSAIDs in humans. To pursue this goal, non-fasted rats were treated for 14 days with

indomethacin 1.5 mg/kg BID by intragastric route, suspended in 1% methylcellulose

and administered in a volume of 0.3 ml/rat. This dose was previously shown to

suppress COX-1 and COX-2 activity in rats by 97% and 98%, respectively [35] and to

reduce inflammation in rat models of adjuvant arthritis [36] and carrageenan-induced

paw edema [37].

The EIR formulation of rifaximin polymorph alpha (R-EIR 50 mg/kg BID) (Alfa

Wassermann SpA, Bologna, Italy) was administered 1 hour before indomethacin for 14

days (suspended in 1% methylcellulose, 1 ml/rat). In a subgroup of animals, treatment

8

with rifaximin was initiated 7 days before starting indomethacin administration and

continued for 14 days until the end of treatment with the NSAID. The dose of 50 mg/kg

BID had been found to protect against enteropathy induced by indomethacin in a

preliminary dose-response study (25, 50 and 100 mg/kg BID). In addition, the selected

rifaximin dose was similar to that employed in previous studies in rats [38,39]. Twenty

four hours after the last dose of test drugs, rats were anesthetized with chloral

hydrate. Blood samples were collected by cardiac puncture from each animal for

hemoglobin measurement. The whole GI tract was excised and samples of jejunum

and ileum were snap frozen in liquid nitrogen and stored at -80°C for subsequent

analysis of myeloperoxidase (MPO), tumor necrosis factor (TNF), malondialdehyde

(MDA) and protein carbonyl levels. Samples of ileum were also processed for the

evaluation of bacterial phyla abundance, as reported below. Other portions of tissues

were fixed in 10% formalin for subsequent evaluation of microscopic damage.

Experimental groups were arranged as follows: Group 1: animals treated with vehicle

(control, n=10); Group 2: animals treated with indomethacin 1.5 mg/kg BID (IND,

n=15); Group 3: animals treated with indomethacin plus R-EIR (50 mg/kg BID) for 14

days (IND+R-EIR14, n=12); Group 4: pretreatment with R-EIR (50 mg/kg BID) for 7 days,

followed by indomethacin co-treatment with R-EIR (50 mg/kg BID) for 14 days (IND+-R-

EIR21, n=12).

2.3 Microscopic assessment of intestinal damage

Histological evaluation of small bowel injury was carried out as previously described

[35]. Upon removal, the small intestine was immediately injected with 10% formalin

and left in the same fixative solution. After 30 min, it was opened along the anti-

mesenteric border, cleaned of its fecal contents, and fixed again in 10% formalin for

24 h. In order to rule out any bias, intestinal tissue samples were taken in accordance

to the following procedure: the full length of small intestine was measured; in the

9

jejunum, two specimens of 1.5-2 cm were taken 5 cm far from both the proximal and

distal end of 81% of the total small bowel length; in the ileum, 1 cm of tissue proximal

to the ileo-cecal valve was discarded and two specimens of 1.5-2 cm were taken at this

edge as well as 5 cm from the first ileum sample.

Sections of jejunum and ileum were embedded into paraffin blocks and cut into 3

consecutive serial 7-8 µm sections. The slices were cut at two different points of the

block: two on the surface and three at a deeper level. Each slice was placed on a glass

slide for staining with haematoxylin and eosin. Histological damage was assessed by

two observers, blind to treatments, according to the score system adopted in our

laboratory [35]. The intestinal damage was scored as reported in Table 1.

Representative pictures, showing the histological appearance of type 1, 2 and 3 lesions

of jejunum and ileum, are displayed in Figure 1.

2.4 Evaluation of tissue myeloperoxidase levels

MPO, as a quantitative index to estimate the degree of intestinal wall infiltration by

inflammatory polymorphonuclear cells, was assessed as described by Fornai et al. [35].

Specimens of small intestinal tissues (30 mg) were homogenized on ice with a polytron

homogenizer (QIAGEN, Milan, Italy) in 0.6 mL of ice-cold lysis buffer (200 mM NaCl, 5

mM EDTA, 10 mM Tris, 10% glycerine, 1 mM phenylmethylsulfonyl fluoride, 1 g/ml

leupeptin and 28 g/ml aprotinin (pH 7.4). The homogenate was centrifuged 2 times at

4°C for 15 min at 1,500 g. The supernatant was diluted 1:5 and used for determination

of MPO concentration by means of an enzyme-linked immunosorbent assay (ELISA)

(Hycult Biotech, Uden, Netherlands). The results were expressed as nanograms of MPO

per milligram of intestinal tissue.

10

2.5 Evaluation of tissue TNF levels

Tissue levels of TNF, a potent inflammatory cytokine, whose intestinal production is

increased dose-dependentely by indomethacin [40] were measured as previously

described [41]. Briefly, samples of jejunum and ileum, collected as reported above,

were weighed, thawed, and homogenized in 0.4 ml of PBS (0.4 ml/20 mg of tissue), pH

7.2 at 4°C, and centrifuged at 10,000g for 5 min. Aliquots (100 μL) of the supernatants

were then used for subsequent assay by means of a commercial ELISA kit (Abcam,

Cambridge, UK). Tissue TNF levels were expressed as picograms per milligram of tissue.

2.6 Evaluation of tissue malondialdehyde levels

MDA concentration in intestinal tissues was determined to obtain quantitative

estimates of membrane lipid peroxidation [35]. For this purpose, intestinal tissue was

excised, weighed, minced by forceps, homogenized in 2 ml of cold buffer (Tris-HCl 20

mM, pH 7.4) using a polytron homogenizer (QIAGEN, Milan, Italy), and centrifuged at

1,500 g for 10 min at 4°C. Aliquots of supernatants were then used for subsequent

assay procedures. Mucosal MDA concentrations were estimated using a colorimetric

assay kit (Cayman Chemical, Ann Arbor, MI, U.S.A.). Results were expressed as nmoles

of MDA per milligram of intestinal tissue.

2.7 Evaluation of protein oxidation levels

Oxidative stress can give rise to protein carbonyl derivatives, via a variety of

mechanisms that include fragmentation and amine oxidation either due to metal

catalysis or by hypochlorous acid. As a consequence, measurement of protein

carbonyls as a marker of tissue injury has become popular [42]. In the rat intestine, the

extent of protein oxidation was estimated using a colorimetric assay kit (Cayman

11

Chemical, Ann Arbor, MI, USA), according to Reznick & Packer [43]. Specimens of

jejunum and ileum (30 mg) were homogenized on ice with a Polytron homogenizer

(Qiagen, Milan, Italy) in 0.6 ml of ice-cold lysis buffer (RIPA buffer containing 1 μg/ml

leupeptin, and 28 μg/ml aprotinin, pH 7.4). The homogenate was centrifuged at 4°C for

10 minutes at 1,600g. Aliquots (200 μL) of supernatants were then used for

subsequent assay procedures. Results were expressed as nanomoles of carbonyl

content per milligram of proteins.

2.8 Assessment of blood haemoglobin concentration

Haemoglobin analysis was performed on blood samples, collected as reported above,

by means of Quantichrom Hemoglobin assay kit (Bioassay Systems, Hayward, CA, USA)

and expressed as g/dL.

2.9 DNA extraction and metagenomic analysis of bacterial population in ileal tissue

DNA was extracted from 250 mg of ileal tissue using the QIAamp DNA Mini kit (Qiagen,

Hildens, Germany) and following the manufacturer’s instructions. The final elution

volume was 200 µl, the DNA concentration was determined by absorbance at 260 nm

(A260), and the purity was estimated by determining the A260/A280 ratio with a

spectrophotometer. Extracted DNA was adjusted to a final concentration of 40 ng/μl.

Genomic DNA from pure bacterial cultures was extracted using Wizard Genomic DNA

Purification Kit (Promega Corporation, Wisconsin, USA) according to the

manufacturer’s instructions.

Total DNA extracted from ileal tissue was analyzed for the mucosal-associated

microbiota through 16S rDNA metagenomics (MiSeq). Metagenomic analysis was

performed with MiSeq, Illumina platform at GenProbio SRL (Parma, Italy). Amplicons of

the V3-V4 bacterial 16S rRNA gene region were obtained and sequenced using the

12

Illumina MiSeq platform, according to Illumina specific protocols. The number of

sequences and operational taxonomic units (OTUs) for each sample were calculated

using Quantitative Insights Into Microbial Ecology software (QIIME,

http://qiime.org/scripts/version1.9.1) and taxonomic classification on Silva v. 119

database was assessed.

2.10 Statistical analysis

Results are presented as mean ± standard error of mean (S.E.M.). The statistical

significance of data was evaluated by one way analysis of variance (ANOVA) followed

by post hoc analysis by Student–Newman–Keuls test, and p values less than 0.05 were

considered significant. All statistical calculations were performed using GraphPad

Prism™ 3.0 software (GraphPad, San Diego, CA, USA).

3. Results

3.1 Mortality Rate

At the end of the treatment period, the group treated with indomethacin displayed a

13.3% mortality rate (Table 2). In both groups of animals treated with rifaximin

(IND+R-EIR14 or IND+R-EIR21) the mortality rate was lower (i.e. 8.3%), albeit not

significantly (Table 2).

3.2 Macroscopic Appearance of the Intestine

Owing to the large extension of intestinal mucosal surface, it was quite difficult to

perform a reliable quantitative estimation of the macroscopic injury elicited by

indomethacin. However, qualitative inspection allowed to detect macroscopic

alterations, including diaphragm-like strictures and multiple ulcerative lesions in

13

animals treated with this NSAID. By contrast, no appreciable macroscopic changes

were evident in intestinal tissues from rats treated with indomethacin plus rifaximin.

3.3 Microscopic assessment of intestinal damage

3.3.1 Effects of indomethacin

In the jejunum and ileum from control animals, microscopic examination did not reveal

any type of lesion (Figure 2 and 3). Administration of indomethacin (1.5 mg/kg BID) for

14 days was associated with the occurrence of type 1, 2 and 3 lesions in both jejunum

and ileum (Figure 2 and 3).

3.3.2 Effects of rifaximin on indomethacin-induced intestinal damage

R-EIR alone did not cause any mucosal injury and did not modify any of study

parameters (data not shown), with the sole exception of bacterial population.

In the jejunum from rats of groups IND+R-EIR14 or IND+R-EIR21 the rates of type I, type

II or type III lesions were significantly decreased or not present at all, as compared with

the group treated with IND alone (Figure 2). Likewise, in the ileum from animals of

groups IND+R-EIR14 or IND+R-EIR21, indomethacin-induced lesions displayed a

significant decrease or were not evident at all, as compared with animals treated with

indomethacin alone (Figure 3).

3.4 Tissue MPO levels

3.4.1 Effects of indomethacin

MPO levels in jejunal specimens excised from control rats were 9.74 ng/mg of tissue.

In animals treated with indomethacin (1.5 mg/kg BID) for 14 days, MPO concentrations

14

were significantly increased (Figure 4A). In the ileum, MPO concentration in control

animals was 7.48 ng/mg of tissue, and the administration of indomethacin (1.5 mg/kg

BID) for 14 days was associated with a significant increase in MPO (Figure 4B).

3.4.2 Effects of rifaximin

In the jejunum from rats of groups IND+R-EIR14 or IND+R-EIR21, tissue MPO levels were

similar to those estimated in control rats, and significantly lower, as compared with

MPO concentrations in jejunal tissues from indomethacin-treated animals (Figure 4A).

Likewise, in the ileum from rats of groups IND+R-EIR14 or IND+R-EIR21, there was a

significant decrease in MPO concentration, as compared with animals treated with

indomethacin alone (Figure 4B).

3.5 Tissue TNF levels

3.5.1 Effects of indomethacin

TNF levels in jejunum from control rats accounted for 2.37 pg/mg of tissue. The

administration of indomethacin (1.5 mg/kg BID) for 14 days was associated with a

significant increase in TNF concentration (Figure 5A). In the ileum, tissue TNF levels in

control rats were 2.09 pg/mg of tissue. In rats treated with indomethacin there was a

significant increment of this cytokine (Figure 5B).

3.5.2 Effects of rifaximin

Rats treated with IND+R-EIR14 or IND+R-EIR21 displayed a significand decrease in tissue

TNF levels both in jejunum and ileum, compared with animals treated with

15

indomethacin alone, showing values not significantly different from those observed in

control animals (Figure 5A and B).

3.6 Tissue MDA levels

3.6.1 Effects of indomethacin

In the jejunum of control rats, MDA concentration was 28.23 nmol/mg of tissue.

Animals treated with indomethacin (1.5 mg/kg BID) for 14 days displayed significant

increments of tissue MDA concentrations (Figure 6A). In the ileum, MDA levels in

tissue samples from control animals were 20.63 nmol/mg of tissue. These levels were

significantly increased by indomethacin (Figure 6B).

3.6.2 Effects of rifaximin

In animals of groups IND+R-EIR14 or IND+R-EIR21 a significant decrease in MDA levels

was observed both in the jejunum and ileum, as compared with indomethacin alone,

with values similar to those estimated in intestinal tissues from control animals (Figure

6A and B).

3.7 Protein carbonyl levels

3.7.1 Effects of indomethacin

The tissue amounts of protein carbonyl in jejunum and ileum form control rats were

18.74 and 19.24 nmol/mg of proteins, respectively. The administration of

indomethacin (1.5 mg/kg BID) for 14 days elicited a significant increase in protein

carbonyl levels both in jejunum and ileum (Figure 7A and B).

16

3.7.2 Effects of rifaximin

In rats treated with IND+R-EIR14 or IND+R-EIR21, there was a significant decrease in

tissue protein carbonyl levels both in jejunum and ileum, with values not significantly

different from those assessed in control tissues (Figure 7A and B).

3.8 Hemoglobin blood levels

Animals treated with indomethacin (1.5 mg/kg BID) for 14 days displayed a significant

decrease in blood hemoglobin concentration, as compared with controls. In rats of

groups IND+R-EIR14 or IND+R-EIR21, hemoglobin levels were significantly higher than

those measured in the group treated with indomethacin alone (Figure 8). Under

rifaximin, however, the hemoglobin levels of indomethacin-treated rats were still

lower than those of control animals, likely because upper GI bleeding was not affected

by the antibiotic.

3.9 Abundance of Proteobacteria, Firmicutes and Bacterioidetes in the ileum

In control animals, the relative abundance of Proteobacteria, Firmicutes and

Bacterioidetes was 10.4, 23.5 and 14.4%, respectively. Treatment with indomethacin

(1.5 mg/kg BID) for 14 days was associated with an increase in the relative abundance

of Proteobacteria and Firmicutes and a slight decrease in Bacterioidetes (Figure 9). In

rats treated with IND+R-EIR14, the relative abundance of such phyla returned toward

those observed in control animals, while R-EIR14 alone did not exert significant effects

(Figure 9).

17

4. Discussion

Results of the present investigation clearly show that rifaximin administration along

with indomethacin protects the small bowel from the damaging effect of this NSAID.

Despite a previous, small study in guinea-pigs [44] reported a protective effect of this

antibiotic on indomethacin-induced intestinal damage, the predictive value of the

experimental model adopted (as well as the drug regimens) is of limited (if any)

relevance to clinical settings.

The present study was conducted on aged rats, since ageing is a risk factor for NSAID-

induced enteropathy [45,46]. Indeed, elderly patients could display altered basal

conditions as a consequence of aging and/or co-morbidities and co-therapies.

Moreover, our model was set up to mirror clinical practice with NSAID chronic

administration in humans, and indeed the present indomethacin regimen caused small

bowel alterations, such as diaphragm-like strictures and multiple erosive lesions,

together with a decrease in blood hemoglobin, known to occur in patients receiving

chronic NSAID therapy [12]. Of note, repeated administration of low dose (i.e. 3

mg/kg) indomethacin, besides mimicking regimens adopted in clinical practice, has

been shown to ensure intraluminal indomethacin concentrations, high enough to elicit

consistent small bowel mucosal damage [47].

The majority of experimental studies on NSAID-enteropathy have used single, high-

dose indomethacin administration to induce mucosal damage. In these models, the

injury pattern consists mainly of multiple, hemorrhagic lesions involving the full

thickness of intestinal wall [48,49,50], a picture that differ greatly from the ones seen

at enteroscopy or video capsule endoscopy in patients taking NSAIDs chronically

[11,12,13]. Most importantly, these acute models of intestinal damage may be not

suitable to evaluate the protective effect of poorly absorbed antibiotics, like rifaximin,

which need at least 7-10 days to consistently affect intestinal bacterial load,

composition and activity [51]. Last but not least, guinea-pig is not considered a

18

reference species to implement a model of NSAID-enteropathy. As matter of fact,

guinea-pigs have been mostly used for in vivo or in vitro studies to evaluate the effect

of indomethacin on intestinal neuromuscular functions [52,53,54], while in vivo

models of enteropathy have been developed in rats or mice, including transgenic

animals. All the above drawbacks strongly limit the translational value of the guinea-

pig acute model and the respective results.

Over 40 years ago, Robert and Asano [55] showed for the first time that germ-free rats

are resistant to the intestinal damaging effect of indomethacin, becoming sensitive

again to such NSAID when mono-contaminated with E. coli. On the other hand, oral

administration of indomethacin is followed by a time-dependent bacterial invasion of

intestinal mucosa (with increase in both aerobic and anaerobic strains) [56], which is

facilitated by the drug-induced increase in intestinal permeability that is also time-

dependent [57]. Enterobacteria then trigger toll-like receptors, whose activation is

followed by an increased mucosal expression of inflammatory cytokines [58]. This

elicits neutrophil recruitment with subsequent release of proteases and reactive

oxygen species (ROS) leading ultimately to mucosal injury [19].

In line with the above-outlined sequence of pathogenic events, we found that

indomethacin did increase mucosal inflammation, as reflected by the increased MPO

and TNF tissue concentrations, thus confirming previous findings [35,59]. Likewise,

lipid and protein peroxidation, mirrored by elevated MDA and protein carbonyl levels,

respectively, also increased after the NSAID administration.

Current prevention strategies to reduce the extent of damage in the upper GI tract are

not effective in the lower GI one. While therapy with celecoxib (a non acidic, COX-2

selective inhibitor) is safer than that with conventional NSAIDs [7], the dose-

dependent increase in the risk of severe CV events associated with long-term therapy

with this class of drugs is of concern in patients with CV risk factors. New alternatives,

including the use of selective and poorly absorbed antibiotics, like rifaximin, probiotics

19

or prebiotics, are therefore important avenues to explore with the aim of correcting

the shift of intestinal microflora towards pro-inflammatory Gram-negative bacteria

[19]. Although current clinical support to this view is still weak, therapeutic

manipulation of luminal microecology is particularly attractive as a physiologic, non-

toxic approach to prevent, if not to treat, NSAID enteropathy.

Being the prevalence of small intestine bacterial overgrowth (SIBO) high in patients on

long-term NSAID therapy [60], the use of rifaximin appears to be an even more

rational choice for prevention of NSAID-enteropathy. Indeed, this poorly absorbed

antibiotic is very effective and widely adopted for treatment of SIBO [61,62], whose

occurrence correlates with the severity of intestinal damage [60].

Our results demonstrate that rifaximin significantly prevents indomethacin-induced

intestinal damage, as documented by macro and microscopic assessment as well as

digestive bleeding. This entero-protective effect was associated with a decrease in

tissue inflammation (i.e. MPO and TNF mucosal concentrations) and oxidative stress

(i.e. MDA and protein carbonyl tissue levels). These findings are consistent with those

obtained in our laboratories with diclofenac-enteropathy [63]. Under similar

experimental conditions, rifaximin was also able to reduce the NSAID-induced increase

in fecal calprotectin as well as to counterbalance the overexpression of TLR-4 and TLR-

2 in the intestinal mucosa [63]. Along the same lines, a video capsule study in healthy

volunteers [64] has recently shown an overall protective effect of rifaximin EIR on

diclofenac-associated intestinal mucosal lesions. Interestingly enough, our

experiments have shown that pre-treatment with rifaximin for 7 days prior to

indomethacin administration does not add further protection in comparison to what

observed when the antibiotic administration was started together with the NSAID,

thus suggesting that a priming effect on enteric microflora is not needed. And indeed,

the pivotal clinical trial [64] confirmed that rifaxmin co-administration with NSAID is

effective in affording intestinal protection. The clinical implications of these findings

are important since pre-administration of a protective agent would delay the start

20

(and benefits) of anti-inflammatory therapy and its co-administration would improve

compliance.

The entero-protective effect of rifaximin likely depends on its broad spectrum of

antibacterial activity, including Gram-positive and Gram-negative bacteria, both

aerobes and anaerobes [32]. More recent experimental studies [65,66] not only

confirmed that rifaximin reduces the total bacterial load, but also modulates bacterial

community composition, increasing the relative abundance of Lactobacillaceae. This

peculiar property was associated with a reduction of intestinal inflammation and

alterations in intestinal permeability [66]. Along the same lines, rifaximin co-treatment

was able – under our experimental conditions – to counterbalance the inflammation as

well as lipid and protein peroxidation induced by indomethacin in the rat small bowel.

Rifaximin-induced changes in bacterial composition my also impact on indomethacin

pharmacokinetics within the intestinal lumen via reduction of the cleavage of its

aglicone by bacterial beta-glucuronidase [67,68]. This would decrease the drug uptake

by enterocytes and, consequently, impair entero-hepatic circulation, which prolongs

the mucosal noxious activity of the NSAID.

The anti-inflammatory activity of rifaximin could be either indirect (i.e. a consequence

of its antimicrobial properties) or direct in nature. Indeed, besides the antibiotic

activity, this drug is endowed with an intrinsic anti-inflammatory activity, which seems

to be class-dependent. Rifamycins indeed inhibit human neutrophil functions [69,70]

and early studies have indeed shown that intra-articular rifamycin is an effective

compound in patients in chronic arthritides, like juvenile rheumatoid arthritis and

ankylosing spondylitis [71]. This anti-inflammatory, which was mirrored in our study by

a significant decrease in MPO (and TNF) tissue levels, could translate into a reduction

of MPO-induced formation of indomethacin reactive metabolites [72].

Recent evidence points out that – besides non-antimicrobial activities – rifaximin

displays also “eubiotic” properties. Indeed, in patients with inflammatory conditions

(like inflammatory bowel disease, colonic diverticular disease or hepatic

21

encephalopathy), rifaximin - while not altering the overall structure of human colonic

microbiota - increased the relative abundance of Bifidobacteria and Lactobacilli

[73,74]. Similarly, our study found that rifaximin was able to counterbalance the

increase in Proteobacteria and Firmicutes abundance induced by indomethacin

administration. Recent investigations [75] found a marked increase in Proteobacteria

in naproxen-treated rats and suggested that these microbial changes may contribute

to NSAIS-induced intestinal damage.

Rifaximin in the prevention of NSAID-enteropathy should clearly used long-term. As for

every drug used in the long-term, safety - besides efficacy - is of paramount

importance. Rifaximin proved to be extremely safe, even when given continuously (at

standard therapeutic doses) for 6 months and its minimal, if any, systemic absorption

(not exceeding 1%) accounts for the adverse event profile, which overlapped that of

placebo [76]. In addition, long-term studies in IBS patients have shown that there were

no clinically relevant changes in bacterial sensitivity to other antibiotic classes, no

emergence of pathogenic bacteria, no occurrence of opportunistic infections, and no

alteration of the overall microbiota [77].

In summary, co-administration of rifaximin with indomethacin can prevent NSAID-

induced small bowel damage. Although the mechanisms need to be further elucidated,

other experimental [63] and clinical [64] studies lend support to the present

conclusions and suggest that targeting enteric bacteria by a poorly absorbed antibiotic

is an attractive therapeutic avenue for the prevention of NSAID-enteropathy [19].

Conflicts of Interest

Corrado Blandizzi has occasionally been involved, as a speaker, in satellite symposia

supported by Alfa Wassermann, the manufacturer of rifaximin.

Carmelo Scarpignato is member of the Advisory Board and of the Speakers’ Bureau of

Alfa Wassermann.

Cecilia Renzulli is an employee of Alfa Wassermann SpA.

The other authors have no conflict of interest to disclose.

22

5. References

[1] Brooks PM, Day RO. Nonsteroidal anti-inflammatory drugs--differences and

similarities. N Engl J Med 1991;324:1716-25.

[2] Tramèr MR, Williams JE, Carroll D, Wiffen PJ, Moore RA, McQuay HJ. Comparing

analgesic efficacy of non-steroidal anti-inflammatory drugs given by different

routes in acute and chronic pain: a qualitative systematic review. Acta

Anaesthesiol Scand 1998;42:71-79.

[3] Aronson JK. Meyler's Side Effects of Analgesics and Anti-Inflammatory Drugs.

Amsterdam: Elsevier; 2009.

[4] Lanas A, Hunt R. Prevention of anti-inflammatory drug-induced gastrointestinal

damage: benefits and risks of therapeutic strategies. Ann Med 2006;38:415-28.

[5] Scarpignato C, Hunt RH. Nonsteroidal anti-inflammatory drug-related injury to the

gastrointestinal tract: clinical picture, pathogenesis, and prevention.

Gastroenterol Clin North Am 2010;39:433-64.

[6] Scarpignato C, Pelosini I. Prevention and treatment of non-steroidal anti-

inflammatory drug-induced gastro-duodenal damage: rationale for the use of

antisecretory compounds. Ital J Gastroenterol Hepatol 1999;31 (Suppl 1):S63–72.

[7] Scarpignato C, Lanas A, Blandizzi C, Lems WF, Hermann M, Hunt RH; International

NSAID Consensus Group. Safe prescribing of non-steroidal anti-inflammatory

drugs in patients with osteoarthritis - an expert consensus addressing benefits as

well as gastrointestinal and cardiovascular risks. BMC Med 2015;13:55.

[8] Silverstein FE, Graham DY, Senior JR, Davies HW, Struthers BJ, Bittman RM, Geis GS.

Misoprostol reduces serious gastrointestinal complications in patients with

rheumatoid arthritis receiving nonsteroidal anti-inflammatory drugs. A

randomized, double-blind, placebo-controlled trial. Ann Intern Med

23

1995;123:241-9.

[9] Laine L, Connors LG, Reicin A, Hawkey CJ, Burgos-Vargas R, Schnitzer TJ, et al.

Serious lower gastrointestinal clinical events with nonselective NSAID or coxib

use. Gastroenterology 2003;124:288-92.

[10] Lanas A, Garcia-Rodriguez LA, Polo-Tomas M, Ponce M, Alonso-Abreu I, Perez-Aisa

MA, et al. Time trends and impact of upper and lower gastrointestinal bleeding

and perforation in clinical practice. Am J Gastroenterol 2009;104:1633-41.

[11] Graham DY, Opekun AR, Willingham FF, Qureshi WA. Visible small-intestinal

mucosal injury in chronic NSAID users. Clin Gastroenterol Hepatol 2005;3:55-9.

[12] Maiden L, Thjodleifsson B, Theodors A, Gonzalez J, Bjarnason I. A quantitative

analysis of NSAID-induced small bowel pathology by capsule enteroscopy.

Gastroenterology 2005;128:1172-8.

[13] Fujimori S, Gudis K, Takahashi Y, Seo T, Yamada Y, Ehara A, et al. Distribution of

small intestinal mucosal injuries as a result of NSAID administration. Eur J Clin

Invest 2010;40:504-10.

[14] Goldstein JL, Eisen GM, Lewis B, Gralnek IM, Zlotnick S, Fort JG. Video capsule

endoscopy to prospectively assess small bowel injury with celecoxib, naproxen

plus omeprazole, and placebo. Clin Gastroenterol Hepatol 2005;3:133-41.

[15] Goldstein JL, Eisen GM, Lewis B, Gralnek IM, Aisenberg J, Bhadra P, et al. Small

bowel mucosal injury is reduced in healthy subjects treated with celecoxib

compared with ibuprofen plus omeprazole, as assessed by video capsule

endoscopy. Aliment Pharmacol Ther 2007;25:1211-22.

[16] Wallace JL. Mechanisms, prevention and clinical implications of nonsteroidal

anti-inflammatory drug-enteropathy. World J Gastroenterol 2013;19:1861-76.

24

[17] Fujimori S, Takahashi Y, Tatsuguchi A, Sakamoto C. Omeprazole increased small

intestinal mucosal injury in two of six disease-free cases evaluated by capsule

endoscopy. Dig Endosc 2014;26:676-9.

[18] Washio E, Esaki M, Maehata Y, Miyazaki M, Kobayashi H, Ishikawa H, et al. Proton

pump inhibitors exacerbate NSAID-induced small bowel injury: A prospective,

double-blind, randomized trial. Clin Gastroenterol Hepatol 2015, in press.

[19] Scarpignato C. NSAID-induced intestinal damage: are luminal bacteria the

therapeutic target? Gut 2008;57:145-148.

[20] Smale S, Tibble J, Sigthorsson G, Bjarnason I. Epidemiology and differential

diagnosis of NSAID-induced injury to the mucosa of the small intestine. Best

Pract Res Clin Gastroenterol 2001;15:723-38.

[21] Bjarnason I, Scarpignato C, Takeuchi K, Rainsford KD. Determinants of the short-

term gastric damage caused by NSAIDs in man. Aliment Pharmacol Ther

2007;26:95-106

[22] Higuchi K, Umegaki E, Watanabe T, Yoda Y, Morita E, Murano M, et al. Present

status and strategy of NSAIDs-induced small bowel injury. J Gastroenterol

2009;44:879-88.

[23] Bjarnason I, Takeuchi K. Intestinal permeability in the pathogenesis of NSAID-

induced enteropathy. J Gastroenterol 2009;44 (Suppl 19):23-9.

[24] Lanas A, Scarpignato C. Microbial flora in NSAID-induced intestinal damage: a role

for antibiotics? Digestion 2006;73 (Suppl 1):136-50.

[25] Scarpignato C, Pelosini I. Rifaximin, a poorly absorbed antibiotic: pharmacology

and clinical potential. Chemotherapy 2005;51 (Suppl 1):36-66.

[26] Rácz I, Szalai M, Kovács V, Regőczi H, Kiss G, Horváth Z. Mucosal healing effect of

25

mesalazine granules in naproxen-induced small bowel enteropathy. World J

Gastroenterol 2013;19:889-96.

[27] Bjarnason I, Hayllar J, Smethurst P, Price A, Gumpel MJ. Metronidazole reduces

intestinal inflammation and blood loss in non-steroidal anti-inflammatory drug

induced enteropathy. Gut 1992;33:1204-8.

[28] Freeman CD, Klutman NE, Lamp KC. Metronidazole. A therapeutic review and

update. Drugs 1997;54:679-708.

[29] Marchi E, Montecchi L, Venturini AP, Mascellani G, Brufani M, Cellai L. 4-

Deoxypyrido[1’,2’:1,2]imidazo[5,4-c]rifamycin SV derivatives. A new series of

semisynthetic rifamycins with high antibacterial activity and low gastroenteric

absorption. J Med Chem 1985;28:960-3.

[30] Jiang ZD, DuPont HL. Rifaximin: in vitro and in vivo antibacterial activity - review.

Chemotherapy 2005;51 (Suppl 1):67-72.

[31] Calanni F, Renzulli C, Barbanti M, Viscomi GC. Rifaximin: beyond the traditional

antibiotic activity. J Antibiot (Tokyo) 2014;67:667-70.

[32] Scarpignato C, Pelosini I. Experimental and clinical pharmacology of rifaximin, a

gastrointestinal selective antibiotic. Digestion 2006;73 (Suppl 1):13-27.

[33] Prantera C, Lochs H, Grimaldi M, Danese S, Scribano ML, Gionchetti P; Retic Study

Group (Rifaximin-EIR Treatment in Crohn's Disease). Rifaximin-extended

intestinal release induces remission in patients with moderately active Crohn's

disease. Gastroenterology 2012;142:473-81.

[34] DuPont HL, Ericsson CD. Prevention and treatment of traveler's diarrhea. N Engl J

Med 1993;328:1821-7.

26

[35] Fornai M, Antonioli L, Colucci R, Pellegrini C, Giustarini G, Testai L, et al. NSAID-

induced enteropathy: are the currently available selective COX-2 inhibitors all the

same? J Pharmacol Exp Ther 2014;348:86-95.

[36] Blackham A, Burns JW, Farmer JB, Radziwonik H, Westwick J. An X-ray analysis of

adjuvant arthritis in the rat. The effect of prednisolone and indomethacin.

Agents Actions 1977;7:145-51.

[37] Amann R, Schuligoi R. Inhibition of carrageenan-induced edema by indomethacin

or sodium salicylate does not prevent the increase of nerve growth factor in the

rat hind paw. Neurosci Lett 2000;278:173-6.

[38] Miglioli PA, Allerberger F, Calabrò GB, Gaion RM. Effects of daily oral

administration of rifaximin and neomycin on faecal aerobic flora in rats.

Pharmacol Res 2001;44:373-5.

[39] Harputluoglu MM, Demirel U, Gul M, Temel I, Gursoy S, Selcuk EB, et al. Effects of

rifaximin on bacterial translocation in thioacetamide-induced liver injury in rats.

Inflammation 2012;35:1512-7.

[40] Bertrand V, Guimbaud R, Tulliez M, Mauprivez C, Sogni P, Couturier D, et al.

Increase in tumor necrosis factor-alpha production linked to the toxicity of

indomethacin for the rat small intestine. Br J Pharmacol 1998;124:1385-94.

[41] Antonioli L, Fornai M, Colucci R, Ghisu N, Da Settimo F, Natale G, et al. Inhibition

of adenosine deaminase attenuates inflammation in experimental colitis. J

Pharmacol Exp Ther 2007;322:435-42.

[42] Dalle-Donne I, Rossi R, Giustarini D, Milzani A, Colombo R. Protein carbonyl groups

as biomarkers of oxidative stress. Clin Chim Acta 2003;329:23-38.

[43] Reznick AZ, Packer L. Oxidative damage to proteins: spectrophotometric method

for carbonyl assay. Methods Enzymol 1994;233:357-63.

27

[44] Ciobanu L, Taulescu M, Prundus R, Diaconu B, Andreica V, Catoi C, et al. Effects of

rifaximin on indomethacin-induced intestinal damage in guinea-pigs. Eur Rev

Med Pharmacol Sci 2014;18:344-51.

[45] Lanas A, Goldstein JL, Chan FK, Wilcox CM, Peura DA, Li C, Sands GH, et al. Risk

factors associated with a decrease >/=2 g/dL in haemoglobin and/or >/=10%

haematocrit in osteoarthritis patients taking celecoxib or a nonselective NSAID

plus a PPI in a large randomised controlled trial (CONDOR). Aliment Pharmacol

Ther 2012;36:485-92.

[46] Watanabe T, Tanigawa T, Nadatani Y, Nagami Y, Sugimori S, Okazaki H, et al. Risk

factors for severe nonsteroidal anti-inflammatory drug-induced small intestinal

damage. Dig Liver Dis 2013;45:390-5.

[47] Nygård G, Anthony A, Khan K, Bounds SV, Caldwell J, Dhillon AP, et al. Intestinal

site-dependent susceptibility to chronic indomethacin in the rat: a morphological

and biochemical study. Aliment Pharmacol Ther 1995;9:403-10.

[48] Gyenge M, Amagase K, Kunimi S, Matsuoka R, Takeuchi K. Roles of pro-angiogenic

and anti-angiogenic factors as well as matrix metalloproteinases in healing of

NSAID-induced small intestinal ulcers in rats. Life Sci 2013;93:441-447.

[49] Abimosleh SM, Tran CD, Howarth GS. Emu oil reduces small intestinal

inflammation in the absence of clinical improvement in a rat model of

indomethacin-induced enteropathy. Evid Based Complement Alternat Med

2013;2013:429706.

[50] Yamamoto A, Itoh T, Nasu R, Nishida R. Sodium alginate ameliorates

indomethacin-induced gastrointestinal mucosal injury via inhibiting translocation

in rats. World J Gastroenterol 2014;20:2641-52.

[51] Testa R, Eftimiadi C, Sukkar GS, De Leo C, Rovida S, Schito GC, et al. A non-

28

absorbable rifamycin for treatment of hepatic encephalopathy. Drugs Exp Clin

Res 1985;11:387-92.

[52] Kubota M, Ito Y, Domae M. Actions of prostaglandins and indomethacin on the

electrical and mechanical properties of smooth muscle cells of the guinea-pig

ileocecal junction. Pflugers Arch 1982;394: 347-54.

[53] Nakahata N, Ono T, Nakanishi H. Possible involvement of a product of the 5-

lipoxygenase pathway in mediation of indomethacin-induced inhibition of

cholinergic transmission in guinea-pig ileum. Eur J Pharmacol 1984;104:133-8.

[54] Shahbazian A, Schuligoi R, Heinemann A, Peskar BA, Holzer P. Disturbance of

peristalsis in the guinea-pig isolated small intestine by indomethacin, but not

cyclo-oxygenase isoform-selective inhibitors. Br J Pharmacol 2001;132:1299-309.

[55] Robert A, Aano T. Resistance of germ free rats to indomethacin-induced lesions.

Prostaglandins 1977;14:331-41.

[56] Takeuchi K, Yokota A, Tanaka A, Takahira Y. Factors involved in upregulation of

inducible nitric oxide synthase in rat small intestine following administration of

nonsteroidal anti-inflammatory drugs. Dig Dis Sci 2006;51:1250-9.

[57] Wright MR, Davies NM, Jamali F. Toxicokinetics of indomethacin-induced

intestinal permeability in the rat. Pharmacol Res 1997;35:499-504.

[58] Watanabe T, Higuchi K, Kobata A, Nishio H, Tanigawa T, Shiba M, et al. Non-

steroidal anti-inflammatory drug-induced small intestinal damage is Toll-like

receptor 4 dependent. Gut 2008;57:181-7.

[59] Takeuchi K, Tanaka A, Ohno R, Yokota A. Role of COX inhibition in pathogenesis of

NSAID-induced small intestinal damage. J Physiol Pharmacol 2003;54 (Suppl 4):

165-82.

[60] Muraki M, Fujiwara Y, Machida H, Okazaki H, Sogawa M, Yamagami H, et al. Role

29

of small intestinal bacterial overgrowth in severe small intestinal damage in

chronic non-steroidal anti-inflammatory drug users. Scand J Gastroenterol

2014;49:267-73.

[61] Scarpignato C, Gatta L. Commentary: Towards an effective and safe treatment of

small intestine bacterial overgrowth. Aliment Pharmacol Ther 2013;38:1409-10.

[62] Shah SC, Day LW, Somsouk M, Sewell JL. Meta-analysis: antibiotic therapy for

small intestinal bacterial overgrowth. Aliment Pharmacol Ther 2013;38:925-34.

[63] Colucci R, Ghelardi E, Tirotta E, Piccoli E, Sacco D, Antonioli L, et al. Rifaximin

prevents enteric bacteria alterations and inflammation in a rat model of

diclofenac-induced enteropathy. United European Gastroenterol J 2015; 3(5

Suppl): A314.

[64] Scarpignato C, Dolak W, Lanas A, Matzneller P, Fogli MV, Grimaldi M, et al.

Rifaximin for prevention of NSAID-associated intestinal lesions in healthy

volunteers: a randomized, double-blind, placebo-controlled, video-capsule study.

Gastroenterology 2015;148 (Suppl 1):S68.

[65] Kim MS, Morales W, Hani AA, Kim S, Kim G, Weitsman S, et al. The effect of

rifaximin on gut flora and Staphylococcus resistance. Dig Dis Sci 2013;58:1676-

82.

[66] Xu D, Gao J, Gillilland M 3rd, Wu X, Song I, Kao JY, et al. Rifaximin alters intestinal

bacteria and prevents stress-induced gut inflammation and visceral hyperalgesia

in rats. Gastroenterology 2014;146:484-96 .

[67] LoGuidice A, Wallace BD, Bendel L, Redinbo MR, Boelsterli UA. Pharmacologic

targeting of bacterial β-glucuronidase alleviates nonsteroidal anti-inflammatory

drug-induced enteropathy in mice. J Pharmacol Exp Ther 2012;341:447-54.

30

[68] Saitta KS, Zhang C, Lee KK, Fujimoto K, Redinbo MR, Boelsterli UA. Bacterial β-

glucuronidase inhibition protects mice against enteropathy induced by

indomethacin, ketoprofen or diclofenac: mode of action and pharmacokinetics.

Xenobiotica 2014;44:28-35.

[69] Spisani S, Traniello S, Martuccio C, Rizzuti O, Cellai L. Rifamycins inhibit human

neutrophil functions: new derivatives with potential antiinflammatory activity.

Inflammation 1997;21:391-400.

[70] Spisani S, Traniello S, Onori AM, Rizzuti O, Martuccio C, Cellai L. 3-

(Carboxyalkylthio) rifamycin S and SV derivatives inhibit human neutrophil

functions. Inflammation 1998;22:459-69.

[71] Caruso I. Twenty years of experience with intra-articular rifamycin for chronic

arthritides. J Int Med Res 1997;25:307-17.

[72] Ju C, Uetrecht JP. Oxidation of a metabolite of indomethacin

(Desmethyldeschlorobenzoylindomethacin) to reactive intermediates by

activated neutrophils, hypochlorous acid, and the myeloperoxidase system. Drug

Metab Dispos 1998;26:676-80.

[73] Maccaferri S, Vitali B, Klinder A, Kolida S, Ndagijimana M, Laghi L, et al. Rifaximin

modulates the colonic microbiota of patients with Crohn's disease: an in vitro

approach using a continuous culture colonic model system. J Antimicrob

Chemother 2010;65:2556-65.

[74] Ponziani FR, Scaldaferri F, Petito V, Lopetuso L, Pecere S, Gerardi V, et al. on

behalf of the GutLiver group. Rifaximin treatment increases Lactobacillus

abudance in patients with different gastrointestinal and liver diseases. UEG

Journal 2015;3 (Suppl):A138.

31

[75] Syer SD, Blackler RW, Martin R, de Palma G, Rossi L, Verdu E, et al. NSAID

enteropathy and bacteria: a complicated relationship. J Gastroenterol

2015;50:387-93.

[76] Bass NM, Mullen KD, Sanyal A, Poordad F, Neff G, Leevy CB, et al. Rifaximin

treatment in hepatic encephalopathy. N Engl J Med 2010;362:1071-81.

[77] Pimentel M, Fodor AA, Golden P, Bortey E, Forbes QP. Characterization of Stool

Microbiota in Subjects With IBS-D Receiving Repeat Treatments With Rifaximin in

the TARGET 3 Study. Gastroenterology 2015;148 (Suppl 1):S-655.

32

A

B

Figure 1. Representative pictures showing the histological appearance of type 1, 2 and 3 lesions elicited by indomethacin in the jejunum (a) and ileum (b). Asterisks in the higher magnificated pictures (right) indicate a significant thickening of the submucosal layer with the presence of inflammatory infiltration and vasodilation. In particular, in the jejunal type 3 lesion a marked thickening of the whole gut wall can be observed, requiring a lower (4x) magnification to be fully appreciated.

33

Figure 2. Histomorphometric analysis of damage in the jejunum of rats treated with vehicle (CONTROL), indomethacin (IND, 1.5 mg/kg BID), indomethacin plus rifaximin-EIR (IND+R-EIR14), or R-EIR (50 mg/kg BID) for 7 days followed by indomethacin plus rifaximin (IND+R-EIR21) for 14 days. Each column represents the mean ± S.E.M. obtained from 10-13 animals. *P<0.05; significant difference vs CONTROL; ap<0.05, significant difference vs indomethacin alone.

34

Figure 3. Histomorphometric analysis of damage in the ileum of rats treated with vehicle (CONTROL), indomethacin (IND, 1.5 mg/kg BID), indomethacin plus rifaximin-EIR (IND+R-EIR14), or R-EIR (50 mg/kg BID) for 7 days followed by indomethacin plus rifaximin (IND+R-EIR21) for 14 days. Each column represents the mean ± S.E.M. obtained from 10-13 animals. *p<0.05; significant difference versus control; ap<0.05, significant difference versus indomethacin alone.

35

Figure 4. Effects of vehicle (CONTROL), indomethacin (IND, 1.5 mg/kg BID), indomethacin plus rifaximin-EIR (IND+R-EIR14), or R-EIR (50 mg/kg BID) for 7 days followed by indomethacin plus rifaximin (IND+R-EIR21) for 14 days on tissue myeloperoxidase (MPO) levels in the jejunum (A) or ileum (B). Each column represents the mean ± S.E.M. obtained from 10-13 animals. *p<0.05, significant difference versus control; ap<0.05, significant difference versus indomethacin alone.

36

Figure 5. Effects of vehicle (CONTROL), indomethacin (IND, 1.5 mg/kg BID), indomethacin plus rifaximin-EIR (IND+R-EIR14), or R-EIR (50 mg/kg BID) for 7 days followed by indomethacin plus rifaximin (IND+R-EIR21) for 14 days on tissue tumor necrosis factor (TNF) in the jejunum (A) or ileum (B). Each column represents the mean ± S.E.M. obtained from 10-13 animals. *p<0.05, significant difference versus CONTROL

37

Figure 6. Effects of vehicle (CONTROL), indomethacin (IND, 1.5 mg/kg BID), indomethacin plus rifaximin-EIR (IND+R-EIR14), or R-EIR (50 mg/kg BID) for 7 days followed by indomethacin plus rifaximin (IND+R-EIR21) for 14 days on tissue malondialdehyde (MDA) in the jejunum (A) or ileum (B). Each column represents the mean ± S.E.M. obtained from 10-13 animals. *p<0.05, significant difference versus CONTROL; ap<0.05, significant difference versus indomethacin alone.

38

Figure 7. Effects of vehicle (CONTROL), indomethacin (IND, 1.5 mg/kg BID), indomethacin plus rifaximin-EIR (IND+R-EIR14), or R-EIR (50 mg/kg BID) for 7 days followed by indomethacin plus rifaximin (IND+R-EIR21) for 14 days on tissue protein carbonyl content in the jejunum (A) or ileum (B). Each column represents the mean ± S.E.M. obtained from 10-13 animals. *p<0.05, significant difference versus CONTROL

39

Figure 8. Effects of vehicle (CONTROL), indomethacin (IND, 1.5 mg/kg BID), indomethacin plus rifaximin-EIR (IND+R-EIR14), or R-EIR (50 mg/kg BID) for 7 days followed by indomethacin plus rifaximin (IND+R-EIR21) for 14 days on blood hemoglobin levels. Each column represents the mean ± S.E.M. obtained from 10-13 animals. *p<0.05, significant difference versus control; ap<0.05, significant difference versus indomethacin alone.

40

Figure 9. Relative abundance of Proteobacteria, Firmicutes and Bacterioidetes in ileal tissues obtained from rats treated with vehicle (CONTROL), indomethacin (IND, 1.5 mg/kg BID), indomethacin plus rifaximin-EIR (IND+R-EIR14), or R-EIR (50 mg/kg BID) for 7 days followed by indomethacin plus rifaximin (IND+R-EIR21) for 14 days.

41

Table 1. Microscopic criteria for quantitative estimation of the intestinal injury elicited by indomethacin Type 1 injury

Damage confined to the tunica mucosa

De-epithelization

Significant morphological alterations of the villi

Type 2 injury

Inflammatory infiltration in the submucosa, with thickening of the tunica muscolaris

or serosa

The morphologic framework of tunica mucosa is preserved

Type 3 injury

Damage involves the full thickness of intestinal wall

The morphologic patterns of tunicae are lost

Inflammatory reaction widely extended to the tunica serosa with a significant

increase in thickness

42

Table 2. Mortality rates in the groups of treatment Treatment Dose

(mg/kg/day) N Mortality (%)

CONTROL - 10 (0/10) 0

IND 3 15 (2/15) 13.3

IND+R-EIR14 3+100 12 (1/12) 8.3

IND+R-EIR21 3+100 12 (1/12) 8.3